Template
Search for Similar Keywords
Find semantically related keywords and get search volume and competition. Jonathanboshoff.com
Keyword research in 2025 is less about chasing one “perfect” phrase and more about covering a topic from every angle. Search engines now look for semantic breadth—holonyms, meronyms, hypernyms, and all the other “nyms” that signal true topical authority. Manually collecting those variations and checking their search volume can eat up hours. For a more comprehensive look at streamlining your content workflows, check out our automate content optimization guide. That’s the gap the “Search for Similar Keywords” app fills: it turns a seed list into a ready-to-use table of keyword ideas with volume and competition in seconds.
Walkthrough: How the App Works
Input
You drop a set of seed keywords into a single text box—no formatting rules beyond commas or line breaks.
Step 1 – Generate Semantic Variations (Chat Model)
The app sends your list to a Llama 3.1 model with a tight prompt that asks for 35 new terms: five each of holonyms, meronyms, hypernyms, hyponyms, synonyms, metonyms, and retronyms. Temperature is locked at 0 to keep the output consistent, and the instructions insist on a plain comma-separated list—no headings, no commentary. The result is a dense string of fresh keyword ideas, already cleaned and ready for the next step. This process of constructing a robust keyword pipeline is reminiscent of strategies discussed in our guide on building high-quality AI content pipelines.
Step 2 – Pull Search Metrics (Keyword Search Volume)
The comma list feeds directly into the Keyword Search Volume node. Behind the scenes, this calls an API tied to Google keyword data (language = EN, country = US). It returns three fields per term—keyword, monthly volume, and competition—formatted as a table. Because the node handles batching automatically, it can process dozens of terms in one hit without extra loops or code.
Output
The final table appears in the app’s output pane and can be exported as CSV or copied straight into a content brief, topical map, or clustering workflow.
Customizing This App
Learn more about the process of customizing apps here
Customization Option | Description |
|---|---|
Add intent labels | Adds a classification step to label each keyword as informational, commercial, navigational, or transactional. |
Limit by volume or competition | Filters keywords based on search volume or competition metrics to refine the results before export. |
Switch markets | Enables targeting of different regions by changing the country code or fetching metrics for multiple locales. |
Predict emerging terms | Incorporates future trends by generating and including forward-looking keyword variations. |
Upgrade the used Models. | There's likely new LLM models that have emerged since writing this article, try using them instead of LLaMa 3.1. |
Running It at Scale with Bulk Runs
Create a CSV with a “Keyword List” column—each row contains the seed terms for one topic.
Open Bulk Runs, choose the “Search for Similar Keywords” app, and upload the file.
Map the CSV column to the app input and launch the job. Moonlit spins up each run in parallel.
Download a combined CSV that holds every topic’s variation list and metrics—ready for clustering or pivot-table analysis.
